Drs. Lea Starita (left), Andrew Stergachis (center) and Runjun Kumar: 'This study helps provide a path forward for how best to implement functional data into clinical medicine'
A new survey of nearly 200 genetics professionals indicates there is a strong demand for a comprehensive database with standardized metrics to support using functional evidence in clinical variant interpretation. In addition, respondents voiced the need for clear guidelines on how to handle conflicting functional data in variant classification.
Results of the survey were published in June in the paper, “Insights on improving accessibility and usability of functional data to unlock their potential for variant interpretation,” in the American Journal of Human Genetics. BBI’s Dr. Andrew Stergachis, M.D., Ph.D., and Lea Starita, Ph.D. are the corresponding authors; several BBI and UW Medicine faculty are among the contributors.
“Variants of uncertain significance are one of the major barriers to precision medicine,” said Starita, an Associate Professor in the Department of Genome Sciences at UW Medicine. “Functional evidence has the potential to overcome this barrier entirely.”
“However, it’s extremely difficult for clinicians to find and interpret functional data, which limits its use in precision genomic medicine,” said Runjun Kumar, M.D., Ph.D., an Assistant Professor in the Department of Laboratory Medicine and Pathology at UW Medicine and one of the paper’s co-first authors. “This data is often generated by individual labs and published across hundreds of journals. There’s no central repository for clinical users.”
The survey, which Kumar believes is the largest ever seeking geneticists’ views on data, sought to better understand the needs of the international genomic medicine community to advance the broader use of functional data. The authors aimed to:
- Explore how providers were encountering and reinterpreting variants of uncertain significance.
- Assess providers’ experience with and confidence using functional data.
- Identify perceived barriers to its use.
- Propose solutions for overcoming those barriers.
Respondents consistently raised concerns about the access to functional data and how to judge the reliability of this data.
The paper notes that 94 percent of respondents said that better access to standardized functional data interpretations would improve their usage of functional data in precision medicine.
Moreover, the authors write: “Respondents overwhelmingly reported that they currently use functional data in their practice, but described gaps in their comfort level with different types of functional data. Specifically, respondents are universally least comfortable with interpreting and using data from high-throughput assays for variant classification. As high-throughput functional data are on track to be the dominant functional data type used in variant interpretations, this survey exposes a significant barrier that needs to be addressed.”
“One surprising result of this study was that respondents were skeptical about being told exactly how to use functional data,” Kumar said. “Instead, they wanted the data to be accessible and standardized, so they could evaluate the evidence themselves. I think what the really want is control and autonomy in the interpretation process.”
So, how can all the data be collected and made presentable?
“To address this gap, we are adapting MaveDB, a public repository for datasets from Multiplexed Assays of Variant Effect (MAVEs), to display data needed to support clinical decisions,” said Starita. “And through collaborations with the Atlas of Variant Effects (AVE) Alliance we’re working to establish standardized data models to ensure that the data is interpretable.”
“This study helps provide a path forward for how best to implement functional data into clinical medicine” said Stergachis.
